Location And Selective Tripping Strategy Decision Of Intelligent Distribution Network

With the extensive access of distributed generation and the upgrading and transformation of intelligent distribution network,the network topology of modern distribution network becomes more and more complex,and the ground fault characteristics become different from the traditional distribution network,which is easy to cause the failure of the original fault location scheme,and bring new challenges to the ground fault location of distribution network.With the continuous development of smart distribution network,information communication technology is updated and iterated,and the collection accuracy becomes higher,which provides more advanced technical support for distribution network fault location and identification.Because our country distribution network is mostly neutral non-effective grounding system,when arbitrary two-point ground fault occurred in different phases,there are two kinds of tripping schemes: two faults tripping simultaneously and one fault tripping first.Therefore,it is of great economic and social significance to make a selective trip decision and remove one of the faults according to the expectation to reduce the power failure area.In this paper,the following research has been done in view of the existing problems of fault detection and selective tripping:Firstly,this chapter introduces the smart distribution network model and intelligent distribution terminal model,and proposes the data preprocessing method suitable for smart distribution network.The data source matrix based on three-phase voltage of power grid is constructed,and the rolling time window method is used to analyze the matrix in real time.To solve the problem of missing data in the time window,linear interpolation method is used to fill in the problem,and then the data source matrix is normalized and standardized.Then,the method of fault time determination and fault location of active distribution network based on Pearson correlation is proposed.Using the difference of voltage waveforms before and after the fault,Pearson correlation coefficient was calculated to realize the distinction and recognition of normal operating state and fault state,providing time reference and data support for subsequent fault identification.The correlation between the 1-mode voltage of each node and the ABC three-phase voltage is analyzed,and the combination faults are located by the number of nodes whose location coefficient falls below the threshold.Calculate the correlation coefficient between the 0 mode voltage and the three phase voltage before the fault,and judge the fault phase according to the negative value.The feasibility of this method is verified by simulation,which can quickly and accurately realize the time determination of ground faults,fault location and fault identification,and has certain robustness and resistance to transition resistance.Finally,a selective tripping strategy based on node importance for single-phase ground combination fault is proposed.The structure index and state index of complex network theory are taken as the evaluation index of node importance,and the entropy-AHP model is established to calculate the comprehensive weight.By using the improved TOPSIS of GRA to calculate the selective tripping index,the selective tripping of single-phase grounding combination faults is completed.The method overcomes the limitation of single weight,and the simulation results show that the decision is reasonable and feasible.